Diffusion Tensor Imaging: A Possible Biomarker in Severe Traumatic Brain Injury and Aneurysmal Subarachnoid Hemorrhage?

Background: A great need exists in traumatic brain injury (TBI) and aneurysmal subarachnoid hemorrhage (aSAH) for objective biomarkers to better characterize the disease process and to serve as early endpoints in clinical studies. Diffusion tensor imaging (DTI) has shown promise in TBI, but much less is known about aSAH.

Objective: To explore the use of whole-brain DTI tractography in TBI and aSAH as a biomarker and early endpoint.

Methods: Of a cohort of 43 patients with severe TBI (n = 20) or aSAH (n = 23) enrolled in a prospective, observational, multimodality monitoring study, DTI data were acquired at approximately day 12 (median, 12 days; interquartile range, 12-14 days) after injury in 22 patients (TBI, n = 12; aSAH, n = 10). Whole-brain DTI tractography was performed, and the following parameters quantified: average fractional anisotropy, mean diffusivity, tract length, and the total number of reconstructed fiber tracts. These were compared between TBI and aSAH patients and correlated with mortality and functional outcome assessed at 6 months by the Glasgow Outcome Scale Extended.

Results: Significant differences were found for fractional anisotropy values (P = .01), total number of tracts (P = .03), and average tract length (P = .002) between survivors and nonsurvivors. A sensitivity analysis showed consistency of results between the TBI and aSAH patients for the various DTI measures.

Conclusion: DTI parameters, assessed at approximately day 12 after injury, correlated with mortality at 6 months in patients with severe TBI or aSAH. Similar patterns were found for both TBI and aSAH patients. This supports a potential role of DTI as early endpoint for clinical studies and a predictor of late mortality.

Abbreviations: aSAH, aneurysmal subarachnoid hemorrhageDTI, diffusion tensor imagingFA, fractional anisotropyGOSE, Glasgow Outcome Scale ExtendedTBI, traumatic brain injuryTE, echo timeTR, repetition time.